1. Field of the Invention
The present invention relates to an image processing apparatus that scans both sides of an original during one-time original feed.
2. Description of the Related Art
An image processing apparatus equipped with an auto document feeder (ADF) is in widespread use. The ADF is mounted in a scanner unit of the image processing apparatus, and it automatically feeds originals one by one. In recent years, an image processing apparatus capable of scanning both sides of an original during one-time original feed has been developed.
In some image processing apparatuses, a user can set a scale factor used to scale up/down an image to be scanned. Then, image scanning is performed by controlling a scale factor in a sub-scanning direction based on the set scale factor. In this method, the scale factor in only the sub-scanning direction adjusted by changing the resolution in the sub-scanning direction. Thereafter, interpolate or reduction of scanned image data obtained in this manner is performed to achieve digital scaling. However, this method has a problem that the higher the scale factor set by the user is, the larger the amount of image data to be scanned gets.
In general, in an image processing apparatus, such as a digital copier, the scanned image data is temporarily stored first in a buffer memory. The image data is then compressed and the compressed image data is stored in an image memory. Meanwhile, in an image processing apparatus equipped with an ADF, in a series of operations from when each of originals automatically-fed by the ADF is scanned till when compressed image data is stored in the image memory, if the image memory becomes full while the originals are scanned, no more data can be stored in the image memory. Therefore, until compression of scanned image data during one-time original feed and storage of compressed image data in the compressed image memory are properly completed, the whole scanned image data is stored in the buffer memory.
However, this method has a problem that when both sides of an original are to be scanned, and a high scale factor is set, a large-capacity buffer memory is required, which leads to an increase in cost.
Japanese Patent Application Laid-open No. 2002-314801, for example, discloses a solution to this problem. In a single-sided scan, a scaled image is obtained by using a scaling method of changing a document feed speed of an ADF. On the other hand, in a double-sided simultaneous scan, an amount of image data is estimated from the size of the original to be scanned and a specified scale factor, it is determined if the image data can be stored in a buffer memory, and if the image data cannot be stored in the buffer memory, the original is scanned at the same magnification (i.e., a scale factor of 1), and then, the image data stored in the buffer memory is digitally scaled at the specified scale factor.
However, in the technology disclosed in Japanese Patent Application Laid-open No. 2002-314801, the scaling method may differ between the single-sided scan and the double-sided scan, and thus an image quality may differ between the single-sided scan and the double-sided scan. In this technology, a case where the scaling method differs between the single-sided scan and the double-sided scan is explained below.
The buffer memory shall have available space corresponding to image data of two pages of originals scanned at the same magnification (100% a scaling). First, in a case of a single-sided original, when 200% scaling is specified, a scaling method A is employed. The scaling method A is 200% scaling (performed by a primary resolution converting unit)×100% scaling (performed by a secondary resolution converting unit). Incidentally, a document feed speed is twice as fast as a document feed speed in 100% scaling. Hereinafter, the document feed speed in 100% scaling is referred to as a 1× speed, and a double speed of the 1× speed is referred to as a 2× speed.
On the other hand, in a case of a double-sided original, when 200% scaling is specified, a scaling method B is employed. The scaling method B is 100% scaling (performed by the primary resolution converting unit)×200% scaling (performed by the secondary resolution converting unit). Incidentally, a document feed speed is fixed at the 1× speed.
The primary resolution converting unit performs scaling on image data by changing the document feed speed of the ADF, and obtains a scaled image. For example, to obtain a 200% scaled image, the primary resolution converting unit changes the document feed speed of the ADF to the 2× speed. The secondary resolution converting unit performs scaling on image data with a scaling circuit composed of an electrical circuit without changing the document feed speed of the ADF, i.e., with the document feed speed fixed at the 1× speed.
In general, when the scaling method differs, an image quality of a scanned image also differs. When the scaling method differs between the single-sided scan and the double-sided scan, and especially, when an original is a halftone photo, moire appears on a scanned image, i.e., the effect of the scaling method is significant. On the other hand, when an original is composed mainly of a text, even if the scaling method is switched between the single-sided scan and the double-sided scan, there is little difference in image quality (appearance of moire). In this manner, a degree of the effect is different depending on a type of original.
In addition to the above problems, there is a need for improvement in scan productivity at least in the single-sided scan with the scaling methods switched between the single-sided scan and the double-sided scan. In this case also, in the technology disclosed in Japanese Patent Application Laid-open No. 2002-314801, as the scaling method differs between the single-sided scan and the double-sided scan, an image quality may differ between the single-sided scan and the double-sided scan.
A contact image sensor (CIS), which is sometimes used for a back-sided scan in a double-sided simultaneous scan, accommodates only a scan at the same magnification. Thus, the CIS may not accommodate the scaling method for obtaining a scaled image by changing the document feed speed of the ADF. Furthermore, depending on a configuration of the scanner unit, in the double-sided scan, an original may be scanned at the same magnification constantly.
For example, in the single-sided scan, it is possible to accommodate the scaling by changing the document feed speed of the ADF; however, in the double-sided scan, only a scan at the same magnification is available due to constraints of the CIS. In other words, when 50% scaling is specified, and originals is a mixture of a single-sided original and a double-sided original, the scaling method differs between the single-sided original and the double-sided original.